I came to realize that testing a manikin in a laboratory would not provide anywhere near the results that I would get if I tested real people in the real world -- during real backpacking trips. Manikins are nothing like real people; we move, we perspire and, well, we're human -- we get up and down during the night, put on and take off clothing, and adjust our gear for comfort. So I set out to find a test method that would work, and what I found was a relatively simple, inexpensive, and repeatable method that will help us to build -- and buy -- sleeping bags.
Let's pretend that 1000 bags of a certain model are sold. If 10% of those buyers used the data loggers to test their bags for two nights each, we'd have 200 hundred nights' worth of tests just to get started. I'd say that date might be a far better metric than three tests carried out on manikins on a laboratory bench. And if we profile each of these testers -- and their gear -- we'll begin to see the real-world performance of sleeping bags, baselayers, sleeping pads, and tents.
Here is my introduction to the Sleeping Bag Test Protocol and links to how you can participate.